ACL Reconstruction With Autografts
Weighing Performance Considerations and Postoperative Care
John A. Grant, PhD; Nicholas G. Mohtadi, MD, MSc
THE PHYSICIAN AND SPORTSMEDICINE - VOL 31 - NO. 4 - APRIL 2003
In Brief: Anterior cruciate ligament (ACL) reconstruction is the treatment of choice for patients who experience episodes of instability and a decreased quality of life after ACL rupture. The bone-patellar tendon-bone and hamstring autografts are the current standards for ACL reconstruction. Primary care physicians, especially sports medicine clinicians, are the first-line providers of nonoperative care for patients who have ACL injuries. Care providers need to know the biologic and biomechanic properties of these grafts, clinical indications for each graft, and rehabilitation considerations to appropriately counsel their patients.
Ligament reconstruction has become the treatment of choice for individuals who experience recurrent instability following rupture of the anterior cruciate ligament (ACL). During the constant evolution in ACL reconstruction surgery, various structures used as ACL substitutes have been categorized as autograft (ie, the patient's own tissue),1,2 allograft (ie, tissue from another human donor),1 or synthetic tissue.3-5 Currently, autograft tissue is the most commonly used.1,6-9 The two tendons used most often for autografts are the central third of the patellar tendon from bone to patellar tendon to bone (BPTB) and the semitendinosus tendon (ST) alone or with gracilis (STG).1,7-9 The BPTB graft is harvested with a bone plug on each end (figure 1); the ST and STG are free grafts (tendon only).
Today, hamstring and patellar tendon grafts are used about equally often.10-12 As providers of nonsurgical care for ACL injury, healthcare practitioners should be prepared to counsel their patients regarding ACL reconstruction surgery. When deciding which of these grafts to use in a particular patient, many factors could affect the success of both the surgical procedure and the patient's successful return to activity. The physician and patient must be aware of these variables to make an informed decision about graft selection, and the physical therapist must be informed to guide optimal recovery. These factors include the biologic and biomechanic properties of each graft, the patient's relevant medical history and clinical findings, graft harvest site healing, and graft-specific rehabilitation considerations.
All of the graft tissues currently used for ACL reconstruction are tendons. Tendons and ligaments appear similar under gross observation and simple histologic examination. However, closer examination reveals the hypercellularity of ligaments relative to tendons and differences in the proportion of total collagen, the amount of glycosaminoglycans, and the proportion of reducible collagen cross-links.13,14 Transformation of these characteristics constitutes the major component of the graft remodeling process.
Human studies of the remodeling process are limited to postmortem analysis and patients undergoing either second-look arthroscopies or total knee replacements.15-19 Human graft remodeling takes place over 3 years,18 substantially longer than the 1-year remodeling period that has been demonstrated in animals.13,20-23 In humans, rapid remodeling occurs in the first 10 months, followed by a 2-year maturing period. After this time, the graft is similar to the native ACL in cellularity, vascularity, and collagen organization but is not "normal."18
Human BPTB and ST grafts appear to go through a similar process of intra-articular remodeling16,18,19,24; therefore, when comparing the graft tissues, more interest has been paid to the remodeling that occurs within the osseous tunnels. Studies15,22,23,25 have reported evidence of both histologic and biomechanic incorporation of BPTB and ST grafts within the osseous tunnels of rabbits and sheep 2 to 3 months after implantation. In dogs, bone-to-bone incorporation (as in the BPTB graft) appears to occur about 3 weeks after implantation26,27 but is no different than tendon-to-bone incorporation (as in the ST graft) with respect to failure load 6 weeks after implantation.27
With BPTB grafts, the bone plug and a portion of the grafted tendon are fixed within the osseous tunnels. The limited amount of data available on human subjects undergoing revised ACL reconstruction has demonstrated results similar to animal studies. Bone-to-bone incorporation occurs quickly, with the bone-tendon junction requiring more than 1 year for complete incorporation.28 Given the current paucity of human data, no concise decisions regarding the true remodeling process can be made. Therefore, the available biologic information suggests that excessive graft strain should be avoided for at least 3 months following surgery to allow for successful graft incorporation.
Strength, Loading, and Stiffness
The full range of mechanical and structural properties of BPTB and ST grafts have been studied.6,29-31 Important features when comparing the two grafts are the strength of fixation within the osseous tunnels, the ultimate tensile load (maximum force in tension that the ligament can sustain before failure),32 graft stiffness (resistance to stretch), and response to cyclic loading (repeated bouts of stretching and relaxation). Fixation strength is the initial limiting factor, but as the graft tissue becomes solidly incorporated into the osseous tunnels, the ultimate tensile load, the stiffness of the intra-articular portion of the graft, and the graft's ability to adapt to cyclic loading become the deciding factors in graft success. No data are available to determine which of these three factors is the most important. The stiffness of the graft best correlates with the clinical grading of joint laxity on physical exam.33,34
In the often-quoted study by Noyes et al,6 the central third of the BPTB construct was found to fail at 170% of the native ACL's ultimate load, and the ST and gracilis alone failed at 70% and 49% respectively. Given that the graft tissue weakens after implantation and that the ultimate load declines with age,31,35,36 double-looped ST or gracilis tendons or double-looped gracilis tendons have been used to increase the ultimate load of the complete construct.
Comparing fixation methods, Steiner et al34 showed that the best STG construct was the double-looped STG graft with soft-tissue washer and suture fixation. The strongest BPTB graft was fixed with both interference screws and sutures tied over posts. The ultimate loads were 105% of the native ACL for the STG and 84% for the BPTB grafts. While the STG construct exhibited a higher ultimate load, the BPTB stiffness was much higher, at 76% of the native ACL versus 44% for the STG. The STG graft also allowed three times the elongation of the BPTB graft before failure. These results feed the debate about whether ultimate load or stiffness is the most important factor for graft selection.
Apart from the fixation and midsubstance strength of the grafts, the variable that is yet to be resolved is the cyclic loading that could cause long-term relaxation (elongation) of these grafts. The BPTB graft has been shown to elongate less than the STG graft.31,34 However, it is currently unknown whether or not this indicates long-term potential for decreased laxity. The fixation and ultimate load of double-looped STG grafts appear superior to that of BPTB grafts. However, the discrepancies in the reported stiffness and the unknown long-term relaxation effects preclude an obvious choice of graft based on biomechanic evaluation.
Choosing Hamstring or Patellar Tendon
The clinical variables of time since injury, amount of laxity, relevant medical history, and postoperative occupational and leisure activities must be evaluated when choosing which graft to use for ACL reconstruction.
Time since injury. Some suggest that the time between injury and surgery (ie, acute versus chronic reconstruction) may have an impact on graft choice. Depending on the author, an acute reconstruction has been classified as occurring from 3 weeks to 3 months following injury.12,37-43 However, current thinking about early reconstruction revolves around the resolution of joint edema and the restoration of full knee range of motion rather than a set time frame.39,40,43-46
Tolin and Friedman37 reported greater clinical success with acute (within 3 weeks of injury) as compared with chronic ST reconstructions. In acute reconstructions, they recommend that the hamstring graft is beneficial for patients who have mild-to-moderate laxity and that the BPTB graft is better for patients who have moderate-to-severe laxity. However, subsequent studies by O'Neill12 and Karlson et al38 demonstrated no differences in patient outcomes between acute and chronic ST reconstructions.
Laxity. Patients who have moderate-to-severe laxity preoperatively (eg, young females) may benefit from using stiffer BPTB grafts. Less stiff ST or STG grafts may be better for patients who have only mild preoperative laxity or a higher probability for postoperative stiffness (eg, older patients or those with osteoarthritis). Further research is needed to settle the debate concerning ultimate load versus stiffness outcomes for these patients.
History. Apart from fixation or incorporation and acute versus chronic situations, a hamstring graft may be preferable in a number of other situations. The disruption to the extensor mechanism and physical insult to the patellar tendon with the use of a BPTB graft can lead to patellar tendinitis.47,48 For patients who have a history of patellar tendinitis, use of a hamstring graft may avoid further insult to the patellar tendon. For patients who have a history of patellofemoral pain, the altered extensor mechanism after midtendon removal may increase stresses on the patellofemoral joint, potentially increasing symptoms.49,50
The use of the patellar tendon graft may also be questionable in patients who have a history of Osgood-Schlatter disease. The hypertrophic osteoblast activity in the tibial tubercle may cause extension of osseous tissue into the distal patellar tendon.51,52 The removal of the osseous tissue will leave a defect in the patellar tendon, thereby weakening the graft. McCarroll et al51 reported successful outcomes despite the presence of an ossicle; however, preoperative radiologic examination of the patellar tendon may help determine the size of the ossicle and the subsequent usefulness of the patellar tendon as a graft.
Activities. Patients who are required to kneel extensively as part of their occupation or other activities may have problems with a BPTB graft. This is mainly due to discomfort at the graft harvest site53 (figure 2), but it may also be partly caused by the superficial numbness over and lateral to the incision that results from cutting the infrapatellar branch of the saphenous nerve during graft harvesting.54 Kartus et al54 have reported a two-incision graft harvest technique that alleviates the superficial numbness by avoiding the infrapatellar nerve, but use of this technique has not been reported outside this group in Sweden.
Finally, the graft choice will also influence the type and amount of strength deficit during the first year postoperatively. A BPTB graft will result in a greater quadriceps deficit, whereas a hamstring graft will result in a larger and more prolonged hamstring strength deficit.55-57
Graft Harvest Site Healing
The harvesting of autologous tissue for ACL reconstruction undoubtedly causes local morbidity and disrupts the tissue mechanics of the graft. Harvesting the BPTB graft leaves a 9- to 11-mm wide defect in the inferior patella, patellar tendon, and tibial tubercle. While filling the bony defects with bone meal reamed from the tunnels and closing the tendon defect with sutures might assist healing, no data show this is necessary.58-60 One concern is that closing the tendon defect may shorten the tendon, resulting in patella baja and potentially infrapatellar contracture syndrome.46,61,62 Further studies59,61,63 show that patellar shortening occurs in some patients, whether or not the defect is closed.
Compared with the patellar tendon defect, the healing response after the complete removal of a 20- to 30-cm section of semitendinosus tendon would theoretically be less successful; however, this is not the case. Firstly, the semitendinosus muscle belly does not appear to retract after the removal of its tendon.64,65 Secondly, recent magnetic resonance imaging and sonography studies have demonstrated that the tendon regenerates toward its distal insertion in 2 years.64-67 In most patients, the tendon continues to extend 10 to 30 mm above the normal pes anserine insertion and becomes more normal in appearance by 1 year after harvest.64-66 If the tendon regenerates to insert distal to the knee joint (either on the popliteal fascia or the pes anserine tendon), it can regain function as a knee flexor. Prospective radiologic and histologic studies are required to confirm these case series studies.
Postoperatively, rehabilitation considerations for patients who have a BPTB graft or an ST graft are mainly guided by issues concerning graft harvest site morbidity and graft fixation strength. As demonstrated by Steiner et al,34 graft fixation strength depends more on the method of tissue fixation than on tissue type.
BPTB grafts. The removal of the BPTB graft can cause many postoperative rehabilitation challenges. The joint edema and localized pain that develop in the extensor mechanism after surgery can have reflex inhibitory effects on the activation of the quadriceps mechanism.68 The insult to the extensor mechanism will limit the progression of quadriceps strengthening55,56; therefore, it is important to control swelling and pain in the early postoperative period to minimize the detrimental effects on strength during rehabilitation. Measures to control pain and swelling include frequent cryotherapy (20 minutes every 1 to 2 hours) immediately after surgery and active knee flexion and passive knee extension exercises starting postoperative day 1. Pain medication is given as needed the first week. As the rehabilitation progresses, care must be taken to closely monitor the patient for symptoms of anterior knee pain. Progressing too quickly through the rehabilitation process may irritate the weakened patellar tendon.
Additionally, postoperative swelling and discomfort may affect the patient's range of motion, which may lead to patellar entrapment and decreased scar mobility.46,62 Early control of pain and swelling are important for range-of-motion improvement. While these goals are important regardless of graft choice, early full passive knee extension, full weight bearing, and patellar mobilization with gliding movements are especially important with a BPTB graft to reduce the risk of patellar entrapment.62
ST grafts. For patients who have an ST graft, the rehabilitation considerations are less specific. The two important factors for patients and therapists to consider are the decreased graft stiffness (ie, increased likelihood of elongation when stressed) and the need for a slower progression of hamstring exercises during early rehabilitation.
To avoid excessive strain on the graft, avoiding open-kinetic-chain knee extension exercises through the 0° to 45° range is important. Active flexion and resisted flexion exercises should progress carefully to allow the remaining unattached semitendinosus tendon to heal without undue aggravation. Eccentric hamstring activity (as in the down phase of prone knee flexion-extension exercises) should be instituted with the support of the opposite leg until activity-induced discomfort has decreased and quality neuromuscular control has been established. Once the hamstrings have initially healed, however, emphasis must be placed on restoring hamstring strength.
Finally, attention should be paid to the amount of force generated in quadriceps-strengthening activities. These activities can progress faster in ST-graft patients because there is no insult to the extensor mechanism. The progression of these exercises should be monitored and balanced with the decreased ability of the hamstrings to counteract anterior tibial translation.
Knee range of motion, ligament laxity, and quadriceps and hamstring strength are widely used for postsurgical evaluation.12,38,49,69-74
Range of motion. In compliant patients who have uncomplicated reconstruction procedures, the accepted accelerated rehabilitation programs are successful in avoiding problems with range of motion.70,75,76 When range-of-motion restriction occurs, it appears to be associated with BPTB graft choice. In head-to-head studies,12,49 extension loss has been more prominent in BPTB grafts compared with ST grafts. Flexion loss is less of an issue but is still more common in patients who have BPTB grafts.12,49,69,74 Patellar crepitus and decreased patellar height may be associated with problems regaining full range of motion, but no clear difference between the grafts has been demonstrated.12,49,50,62
Ligament laxity. Measured with the KT1000 (MEDMetric Corp, San Diego), ligament laxity is the most consistently reported evaluation of ACL reconstruction success.12,38,49,69-74 While passive laxity measures have not correlated with the functional ability of a patient,77 laxity provides an important assessment of the graft's patency in the joint. Studies12,69-71,73 have demonstrated that 83% to 92% of BPTB-grafted patients and 82% to 88% of ST-grafted patients have acceptable joint laxity 2 years after surgery. Interestingly, in ST-grafted patients, Noojin et al71 demonstrated a long-term trend toward a higher percentage of men (88%) having acceptable laxity as compared with women (79%).
Strength deficits. Graft-specific strength deficits have been reported up to 1 year postoperatively.55,56 However, longer term studies have not shown any differences in quadriceps and hamstring strength deficits between BPTB and ST grafts.12,38,49 Further research is required to determine if the return of strength is related to the regeneration of the harvested tissue.58,64-66,78
Measures of Success
Subjective patient reports of surgery success are widely used but difficult to quantify and pool into a consensus. Global scores, such as the International Knee Documentation Committee (IKDC) questionnaire and Lysholm scales,79-84 that incorporate objective findings (range of motion, laxity, functional performance) and subjective patient reports of satisfaction have been published and are gaining popularity. The wide variety of both published and author-developed satisfaction surveys makes it difficult to compare outcomes across studies. Little difference was seen more that 2 years after surgery between patients who had reconstruction with BPTB or ST grafts in studies that used either IKDC or Lysholm scales. 12,38,71
Given the nonsignificant findings published in the literature, a 2001 meta-analysis11 comparing the two grafts demonstrated that the BPTB graft resulted in significantly more patients returning to their preinjury sport and having acceptable long-term ligament laxity. This meta-analysis also showed a trend for better outcomes with a BPTB graft in all the other clinical stability measures, and that the ST and BPTB grafts were equally successful in range of motion, postoperative complications, and graft failure. Since this publication, three more studies85-87 have been published on this topic. Two studies85,86 appear to strengthen the meta-analysis; the third87 demonstrates a preference for the ST graft.
Published studies show the BPTB graft to be superior in clinical stability, but strength outcomes have not been routinely evaluated in these studies. Quadriceps strength tends to lag in patients who have a BPTB graft, and hamstring strength trends to lag in patients who have an ST graft. The effects of these strength deficits, especially over the long term, need to be evaluated with functional tests (eg, stair hopping or one-legged hopping) to completely assess the most successful graft choice.
Weighing the Evidence
Considering the vast array of published research, it is astounding that so many questions remain to be answered regarding graft selection and long-term outcome after ACL reconstruction. Evidence demonstrates that both graft types mature in a similar fashion. The time to maturity is still unknown, as is the exact process of incorporation within the bony tunnels. Biomechanically, the method of fixation is more important in the early healing phase. It is currently unknown whether ultimate failure load, stiffness, or cyclic loading characteristics of the graft are more important for long-term success.
Due to the increased risk of patellar symptoms, ST grafts are likely a better choice for patients who have a history of anterior knee dysfunction (eg, patellofemoral pain, patellar tendinosis). These grafts may also be preferred by patients who kneel regularly. Rehabilitation considerations are specific to graft harvest site and fixation technique. Physical therapists should guide patients to progress within their healing abilities, keeping in mind the biologic progression of healing and the course of biomechanic weakness.
The BPTB graft has been shown to provide a more stable knee in the long term. The preferred graft for the return of functional ability and overall patient satisfaction is still up for discussion. To date, there is still no consensus on the comparative success of the two graft types in these outcomes, which may be the most important to the patient. In most cases, it boils down to the surgeon's preference and skill.
Dr Grant is an MD candidate, and Dr Mohtadi is a clinical associate professor and staff orthopedic surgeon at the University of Calgary Sport Medicine Centre in Calgary, Alberta, Canada. Address correspondence to John A. Grant, PhD, University of Calgary Sport Medicine Centre, 2500 University Dr NW, Calgary, Alberta, Canada, T2N 1N4; e-mail to [email protected].
Disclosure information: Drs Grant and Mohtadi disclose no significant relationship with any manufacturer of any commercial product mentioned in this article. No drug is mentioned in this article for an unlabeled use.